skills/vercel-labs/lat.md/lat-md/Gen Agent Trust Hub

lat-md

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFE
Full Analysis
  • [Documentation Standards]: The skill defines a structured approach to project documentation, emphasizing rationale, design decisions, and test specifications. This methodology supports project maintainability and does not introduce security vulnerabilities.
  • [Validation Tooling]: The instructions encourage the use of the lat check tool to verify the integrity of wiki links, section structures, and code references. This automated validation helps ensure documentation consistency and link safety.
  • [External Source Management]: The skill references the management of external sources through the lat CLI tool. It specifically highlights reviewing pinned commits and following checkout suggestions, which aligns with security best practices for handling external dependencies.
  • [Indirect Prompt Injection Surface]: As the skill involves processing and generating documentation that serves as agent context, it is a surface for indirect prompt injection. However, the requirement for specific Markdown structures, leading paragraphs, and validation via lat check serves as a natural mitigation against unstructured instruction injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 07:41 AM
Security Audit — agent-trust-hub — lat-md